Isolation and Some Properties of Human Metallothionein*

presence of mercury in human metallothionein was traced to the use of organic mercurials during therapy of patientswhose kidneys served for isolation ...
0 downloads 0 Views 969KB Size
BIOCHEMISTRY

Isolation and Some Properties of Human Metallothionein* Pablo Pulido,t Jeremias H. R. Kagi,: and Bert L. Vallee

The most highly purified preparations of metallothionein isolated from human renal cortex obtained so far contain 4.2 % cadmium, 2.6 % zinc, 0.5 mercury, and 0.3% copper, maximally a total of 8.9 g-atoms/molecular weight of 10,500 f 1050, thus being quite similar to the equine protein [Kagi, J. H. R., and Vallee, B. L. (1961), J . Biol. Chem. 236, 24351. The presence of mercury in human metallothionein was traced to the use of organic mercurials during therapy of patients whose kidneys served for isolation of the protein. For each metal atom of human metallothionein three mercapto groups appear to be available for

binding, quite similar to the equine protein. The tot21 metal-binding capacity is accounted for by 26 silvertitratable sulfhydryl residuesimole, and in accord with this, preliminary amino acid analyses indicate that cysteinyl residues represent a minimum sulfur content of 8.1 %. Human and equine metallothionein both exhibit a similar absorption band at 250 mp, characteristic of cadmium mercaptide chromophores. Both proteins also exhibit closely similar rotatory dispersion. There is a large positive Cotton effect centered at 254 mp, due to the asymmetric binding of cadmium to multiple sulfur ligands.

ABSTRACT:

M

etallothionein was first isolated from equine renal cortex by Margoshes and Vallee (1 957). Further purification showed it to contain 5.9% cadmium and 2.2 zinc. The metal-free protein, thionein, contains 16.3% nitrogen and 9.3% sulfur (Kagi and Vallee, 1961). A very similar metalloprotein has now been isolated from the renal cortex of the human kidney through the application of an improved isolation procedure in which gel filtration substitutes for salt fractionation, used previously (Kagi and Vallee, 1960). While cadmium and zinc constitute most of the metal content of the equine protein, some preparations of human metallothionein also contain substantial quantities of mercury. The occurrence of this metal was traced to the therapeutic use of mercurial diuretics in the patients from whom kidneys were obtained at autopsy. In the best preparations obtained thus far, the total metal content of human metallothionein amounts to 8.9 g-atoms/10,500 mol wt. There are three silver-reactive mercapto groupsjg-atom of metal bound. Both values are close to those previously found in the equine protein.

necessary precautions to avoid metal contamination have been described (Thiers, 1957). Solutions were stored in polyethylene containers at 4". All dialyses were performed in cellulose casings (Visking-Nojax, 27/32), precleaned, and treated as previously described (Kagi and Vallee, 1960). pH measurements were made potentiometrically at 23 + 2" either with a Beckman Model G or with a Radiometer (Copenhagen) Model 25 pH meter. Conductivity determinations were performed with a Radiometer conductivity meter at 23 f 2". Protein concentrations were determined gravimetrically after precipitation with trichloracetic acid and drying at 104" (Hoch and Vallee, 1953). The absorbance at 250 mp served as alternate method for estimating the concentration of metallothionein (Kagi and Vallee, 1961). The absorptivity was determined to be 6.8 ml mg-l cm-' for a preparation of highly purified human protein. A Beckman DU spectrophotometer was used throughout. Metal analyses were carried out by emission spectrography (Vallee, 1955). Cadmium, zinc, and copper were also determined by atomic absorption spectrophotometry after dilution of chromatographic fractions in water (Fuwa and Vallee, 1963; Fuwa et al., 1964; Pulido et al., 1966). Mercury was measured by atomic absorption spectrophotometry (Cobain, 1965; K. Fuwa, W. A. Cobain, and B. L. Vallee, unpublished data). Reactive mercapto groups were determined by amperometric titration with Ag+ (Benesch et a/., 1955) in a supporting electrolyte solution (Tris, KCI, HNOe) at pH 7.5 (Hoch and Vallee, 1960). Gel Filtration Clwomatography. Gels were prepared by swelling in metal-free distilled water or dilute Tris buffer for at least 10 days. The following gels were employed : Sephadex (Pharmacia, Uppsala, Sweden),

Materials and Methods Reagents and Glassware. Analytical grade chemicals and metal-free water were used throughout. The preparation of metal-free water, metal-free buffers, and the

1768

* From The Biophysics Research Laboratory, Department of Biological Chemistry, Harvard Medical School, and the Division of Medical Biology, Peter Bent Brigham Hospital, Boston, Massachusetts. Receiced March 3, 1966. This work was supported by a grant-in-aid (HE-07297) from the National Institutes of Health, Education and Welfare, and by a grant-in-aid from The Nutrition Foundation. t Recipient, Lederle International Fellowship. $Investigator of Howard Hughes Medical Institute.

P A B L O

PULIDO, JEREMIAS

H.

R.

K:iGI,

A N D

BERT

L.

VALLEE

VOL.

5 , N O . 5, M A Y 1 9 6 6

Figure 1: Purification of Human Kidney Metallothionein. Discard fractions

Main-line fractions Cortex

Residue I +-

homogenization extraction by phosphate butrer centrifugation

J. Supernatant I

Residue I1 +

I ++

J.

ethanol chloroform centrifugation

Supernatant I1 Residue 111 f-

dialysis us. water, pH 7.0 cell trifugation

.1

Supernatant 111 Fraction IVa and + metal-free fractions

1

Sephadex G-50 filtration 0.001 M Tris, PH 8.6

Fraction IV Fraction Va and + metal-free fractions

I J.

TUBE NUMBER. ( 5 m t )

2 : Chromatography of human kidney metallothionein on Sephadex G-50. Supernatant 111 (3.1 g), in 47 ml of 0,001 M Tris, pH 8.6, was chromatographed on a 56 X 4 cm (540 ml) bed of Sephadex G-50 fine beads at a flow rate of 25 ml/hr, 4". Cadmium (I) absorbances at 250 ( 0 )and 282.5 mp (0)were measured. Fractions ( 5 ml) were collected. As indicated by the arrows, the cadmium-rich fraction IV was pooled and lyophilized. The balance was discarded. FIGURE

Sephadex (3-75 tiltratioti 0.001 M Tris, pH 8.6

Fraction V

I Metaimfreefractions +--

polyacrylamide P-20 filtration 0.001 M Tris, pH J. 8.6

Fraction VI Porath column electrophoresis 0.01 hi Tris, pH

Metallothionein (components A-C)

G-25 (Lot 4370) fine beads; G-50 (Lot 4120) fine beads; G-75 (Lot 5225); and polyacrylamide gel, P-20 (BioRad, Lot 2703). Fine particles were eliminated by decantation and the absorbent beds were packed at room temperature by pouring a thin slurry of gel particles in buffer solution into glass columns, partially filled with buffer. Slurry was added until the desired bed height was obtained. A flow rate between 15 and 25 ml/hr was maintained. Preparative column electrophoresis was performed in a Porath column, 2.9 X 30 cm (LKB Model 3340 C), packed with ethanolized cellulose and equilibrated with 0.01 M Tris-HC1, pH 8.6. The column was prepared and the sample applied as described by Porath and Hjerten (1962). Runs (19 hr) using 525 v, 6 ma were performed with the column cooled to 10 i 1 ". Disk gel electrophoresis was performed on samples containing 0.1-0.2 mg of protein (Davis, 1964). Molecular weights were estimated by gel filtration (Andrews, 1965), utilizing a 148 X 1 cm Sephadex G-75 column, equilibrated with 0.04 M Tris-HC1, pH 8.1, at 23 + 2". Protein samples, 1-3 mg in 1 ml, were applied to the column and their elution volumes were estimated by spectrophotometric measurement on 2-ml eluate fractions. The void volume, Vo, was determined after every second run with Blue Dextran

(Pharmacia, Uppsala, Sweden), a polysaccharide with an average molecular weight of 2 X 106. Absorption spectra were determined with a recording spectrophotometer (Cary Model 15) using 1-cm lightpath quartz cells. Optical rotation was measured with a Cary Model 60 recording spectropolarimeter over the range 220-500 mp at 23 + 2"; cells with fused quartz end plates and 2-20-mm path length were used. The slit width of the instrument was programmed to yield constant energy over the entire spectral range. Measurements were performed with protein concentrations varying from 0.5 to 0.05 mg/ml to eliminate the possibility of spurious Cotton effects (Urnes and Doty, 1961). Base lines were recorded in the same cells containing buffer only. Molar rotation with respect to cadmium was calculated from the equation [MI: = [a]:MWjlOO where [a]: = specific rotation and MW = minimum molecular weight calculated for 1 g-atom of cadmium present in the protein. Isolation of Human Metallothionein Human kidneys were obtained from patients of both sexes whose ages ranged from 18 to 71 years and who had died without exhibiting primary renal disease. The organs were shown to be free of pathological changes both by gross and histological features. The kidneys were frozen as soon as possible and stored at -20". Unless otherwise stated all operations were carried out at 4". For each preparation 12-14 kidneys are thawed overnight, rinsed with water, the capsule removed, and the cortex separated from the medulla. The kidney cortex is cut in small strips and homogenized in an electric blendor for 20 sec. The procedure is an extension of that previously employed for equine metallothionein (Kagi and Vallee, 1960) (Figure 1). The homogenate of kidney cortex is extracted with an equal volume of 0.05 M sodium phosphate buffer,

1769

H U M A N METALLOTHIONEIN

BIOCHEMISTRY

ll

1

1

I 0

40 80 TUBE NUMBER. (5m.!)

120

3 : Chromatography of human kidney metallothionein on Sephadex G-75.Fraction IV (616 mg) in 15 ml of 0.001 M Tris, pH 8.6,was chromatographed on a 127 X 2.2 cm (490 ml) bed of Sephadex G-75,at a flow rate of 18 ml/hour, 4".Cadmium (m) absorbances at 250 (o), and 282.5 mp (0) were measured. As indicated by the arrows, the 5-ml fractions were pooled to yield cadmium-rich fraction V which was lyophilized. Fraction V-a was discarded. FIGURE

1770

pH 7.0, for 2 hr with continuous stirring. To remove heavy particles the slurry is centrifuged in an International refrigerated centrifuge at lOOOg for 60 min to yield supernatant I. With constant stirring 1.2 parts of 95 ethanol, followed by 0.095 part of chloroform, prechilled to -20", are added dropwise to the turbid supernatant I. The massive reddish and rubbery precipitate, residue 11, is separated by centrifugation at lOOOg for 1 hr. The resultant clear yellow supernatant I1 is freed from organic solvents and salt by overnight dialysis cs. two changes of water, adjusted, and maintained at pH 7.0 by the addition of 0.01 M sodium hydroxide. The fine precipitate formed on dialysis, residue 111, is removed by centrifugation at 32,600gin a Spinco Model L preparative ultracentrifuge to yield supernatant I11 (approximately 1500 ml) which is lyophilized. Fractionation is continued by chromatography on a series of gels having varying exclusion limits. The elution of metallothionein is monitored by measuring both the cadmium content and the absorbance, at 250 and 282.5 mp of the fractions. Tris is added to supernatant I11 to yield a final concentration of 0.001 M, pH 8.6, and the resultant solution is passed through a 56 X 4 cm Sephadex G-50 column at a flow rate of 25 ml/hr. The molecular weight exclusion limit of the gel was 10,000 as measured with dextran (Figure 2). The major fraction of the total cadmium content is eluted just after the first major absorbance peak, and occurs in the position of partially excluded material. The material showing close correspondence between the cadmium content and the absorbance at 250 mp is pooled and lyophilized (fraction IV). Both fraction IVa, containing a minor amount of cadmium associated with higher molecular weight material, and the second major absorbance peak, containing low molecular weight material, are discarded.

PABLO

PULIDO,

JEREMIAS

H.

R.

K.XGI,

A N D

BERT

TUBE NUMBER. ( 5 r n l )

4: Electrophoresis of human kidney metallothionein in Porath column. Fraction VI (115 mg) in 14 ml of 0.01 M Tris-HC1, pH 8.6, was separated in a 2.9 X 30 cm cellulose bed (285 ml) at IO", 525 v, 6 ma. After 19 hr the material was displaced with buffer and collected in 5-ml fractions at a flow rate of 0.2ml/ min. Cadmium (D) and zinc (V) absorbances at 250 ( 0 ) and 282.5 mp (0) were measured. Fractions were pooled as indicated by the arrows to yield components A-C. FIGURE

Fraction IV dissolved in 0.001 M Tris, pH 8.6, is rechromatographed on a 127 X 2.2 cm Sephadex G-75 column, with an exclusion limit of 50,000 mol wt, at a flow rate of 18 ml/hr (Figure 3). Approximately 75% of the total cadmium applied to the column is recovered in a single peak in the position of nonexcluded, lighter material. This cadmium-rich fraction is pooled and lyophilized (fraction V), while the heavier and excluded material (fraction Va) containing a relatively small amount of cadmium is discarded. Fraction V is purified further by passage through a 127 X 2.2 cm column of polyacrylamide gel (P-20), with an exclusion limit of 20,000, equilibrated with 0.001 M Tris, pH 8.6, at a flow rate of 15 ml/hr. The resultant cadmium-containing peak is well separated from two peaks of heavier and lighter impurities. The combined eluate for the cadmium-containing peak is lyophilized to constitute fraction VI or crude metallothionein. Further purification of fraction VI is performed by column electrophoresis. After the buffer has been exchanged for 0.01 M Tris-HC1, pH 8.6,on a 50 X 1 cm Sephadex G-25 column, fraction VI is applied to the Porath column. After 19 hr of electrophoresis, 5-ml fractions are eluted at a flow rate of 0.2ml/min. Figure 4 shows the resultant elution diagram in which one minor and two major components are observed moving toward the anode. These are designated at components A-C of metallothionein. These components were also demonstrated by disk gel electrophoresis. Further studies were confined to components B and C which accounted for more than 90 % of the cadmium applied to the column.

L.

VALLEE

VOL.

5,

5,

NO.

MAY

1966

Metal Content of Fractions Attending Purification of Metallothionein:

TABLE I :

~~~

Cortex Buffer extract Supernatant I1 Fraction IV Fraction V Fraction VI Metallothionein component B Metallothionein component C

~~~

~

140 130 2,510 5,650 10,100 29,500 42,200 38,500

520 400 130 350 . , . r , . . 950 840 7,520 2,300 1,610 , . , , . . 420 1,470 6,120 . . . . . . . . . 2,880 1,160 10,500 5,300 , , , , . . 28,400 17,200 3,750 1,710 1,120 nd . . . 26,700" 5,380 2,540 . . . . . .

...

19,9O0"4,760 4,600

0.8 1.3 nd 3.7 15 nd . . .

3.4 25 6.1 22 41 28